CN-122017597-A - Cell module testing method and device and electronic equipment

CN122017597ACN 122017597 ACN122017597 ACN 122017597ACN-122017597-A

Abstract

The application relates to a cell module testing method, a device and electronic equipment, wherein the method constructs a cell data model according to experimental data and historical operation data of a target cell in different working stages, and further maps the cell data model into an equivalent circuit model to generate a cell equivalent circuit model for simulation; and the control cell module equipment is used for outputting the voltage, current and charge state change of the analog cell according to the cell equivalent circuit model, collecting output end data in real time in the simulation process, and carrying out dynamic correction and parameter adjustment on the cell data model and the equivalent circuit model so as to realize accurate simulation of the dynamic behavior of the target cell under different charge and discharge working conditions. According to the application, data-driven modeling and equivalent circuit modeling can be fused, and a dynamic adjustment and model correction mechanism is introduced in the simulation process, so that the cell module result can be continuously close to the real state of the target cell along with the change of working conditions.

Inventors

DAI LIUXING
ZHANG JINCHUAN
WANG QINGWEI
LI YUE
SONG LIUHUAN

Assignees

固势(苏州)科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260301

Claims (10)

1. A cell model testing method with fusion regulation function, characterized in that the method comprises: generating multidimensional cell characteristic parameters according to experimental data and historical operation data of a target cell in different working stages, and constructing a cell data model capable of reflecting dynamic behaviors of the cell by adopting a machine learning method or a statistical modeling method according to the multidimensional cell characteristic parameters; mapping the electric core data model into an equivalent circuit model and generating an electric core equivalent circuit model by setting preset circuit parameters; the control cell module equipment outputs the simulated cell voltage, current and charge state change according to the cell equivalent circuit model so as to realize the real-time simulation of the target cell under different charge and discharge working conditions, And in the real-time simulation process, acquiring preset data of the output end of the cell simulation equipment in real time, and carrying out dynamic correction and parameter adjustment on the cell data model according to the preset data acquired in real time so as to enable the cell simulation equipment to truly simulate the application characteristics of the target cell.
2. The method of claim 1, wherein the generating the multi-dimensional cell characteristic parameters according to the experimental data and the historical operation data of the target cell in different working phases and constructing the cell data model capable of reflecting the dynamic behavior of the cell by using a machine learning method or a statistical modeling method according to the multi-dimensional cell characteristic parameters comprises: obtaining experimental data and historical operation data of a target battery cell in different working phases, wherein the experimental data comprise voltage, current, temperature and state change data of the target battery cell in different charge states, and the historical operation data comprise voltage, current, temperature and state change data of a historical record of the target battery cell; After the experimental data and the historical operation data are obtained, carrying out data denoising, abnormal data removing, time alignment and normalization processing, and extracting multidimensional characteristic parameters for representing voltage change rate, current response characteristics, temperature change characteristics, charge state and internal resistance change of the battery cell state; According to the multidimensional electric core characteristic parameters, a machine learning method or a statistical modeling method is adopted to construct an electric core data model, wherein the electric core data model is trained by taking the multidimensional electric core characteristic parameters as model input and taking target electric core output voltage, current and charge state change as model output; the mapping the electric core data model to an equivalent circuit model and generating an electric core equivalent circuit model by setting preset circuit parameters comprises the following steps: and determining circuit parameters of resistance, capacitance and voltage source in the equivalent circuit model according to preset characteristics reflected in the electric core data model, and further constructing the electric core equivalent circuit model for real-time simulation.
3. The method of claim 1 or 2, wherein the cell module comprises a microprocessor, a digital-to-analog conversion circuit, an acquisition circuit, an operational amplifier buffer, a signal comparator, a signal integrator, a power operational amplifier circuit, a sampling resistor, a communication interface, a Sense interface, and a VBAT power output interface.
4. The method of claim 3, wherein the controlling the cell die apparatus to output analog cell voltage, current, and state of charge changes according to the cell equivalent circuit model comprises: The control microprocessor generates a corresponding digital control signal according to a target voltage value, a current value and a state of charge parameter set in the cell equivalent circuit model and sends the digital control signal to a digital-to-analog conversion circuit; The digital-to-analog conversion circuit is controlled to convert the digital control signal into a continuously-changing analog voltage signal and perform stabilization and buffering treatment on the analog voltage signal through the operational amplifier buffer; The control signal comparator compares the buffered analog voltage signal with the actual output voltage signal acquired by the load end through the Sense interface in real time to obtain a deviation signal between the target output signal and the actual output signal, and The control signal integrator performs integration processing on the deviation signal and sends the integrated control signal to the power operational amplifier circuit, and then the corresponding analog cell voltage and current are output through the power operational amplifier circuit.
5. The method according to claim 4, wherein the step of acquiring the preset data of the output end of the cell simulation device in real time and dynamically correcting and adjusting parameters of the cell data model according to the preset data acquired in real time during the real-time simulation includes: The sampling resistor is controlled to collect analog signals at the output end of the analog cell in real time, the collected analog signals are processed into a preset input range through the signal conditioning circuit, and then the conditioned analog signals are converted into digital signals and the digital signals are transmitted to the microprocessor; Controlling the microprocessor to determine the operation condition of the target battery cell in the current control period according to the analog signal and acquiring predicted data matched with the operation condition from a battery cell data model; Controlling the microprocessor to compare the digital signals with the predicted data item by item, and calculating corresponding output voltage deviation values, output current deviation values and state of charge deviation values; Controlling the microprocessor to judge whether the target output parameter calculated by the cell equivalent circuit model in the current control period deviates from a preset threshold value according to the deviation value, if so, judging that the target output parameter deviates from the actual behavior of the target cell, and executing the dynamic correction and parameter adjustment processes; performing model parameter correction, feature mapping relation correction and/or model weight correction on the cell data model according to the deviation value, and And mapping the corrected cell data model into a cell equivalent circuit model again, synchronously updating related circuit parameters in the cell equivalent circuit model, and further recalculating the change of the voltage, the current and the state of charge of the analog cell and updating the output parameters of the cell module equipment.
6. The method of claim 5, wherein during the real-time simulation, the method further comprises: Judging whether the dynamic correction of the cell data model and the cell equivalent circuit model achieves a preset effect or not according to a preset error threshold and a convergence criterion; if yes, the control microprocessor stops adjusting parameters of the cell data model and the cell equivalent circuit model and applies the current parameters as stable parameters to a subsequent control period; If not, the control microprocessor continues to execute the dynamic correction process according to the error evaluation index, and performs incremental adjustment on the parameters until the convergence criterion is met or the preset correction times are reached.
7. The method of claim 1 or 5, wherein during the real-time simulation, the method further comprises: Setting target state of charge parameters and equalization strategy parameters of a plurality of cell module cell modules, wherein the equalization strategy parameters comprise one or more of active equalization parameters or passive equalization parameters; collecting output voltage, current and state of charge parameters of the plurality of cell module devices and calculating state difference values among different cell module devices; Controlling the plurality of cell module devices to adjust the output voltage or the output current in the simulation process according to the equalization strategy parameters so as to form voltage differences or electric quantity differences among different cell module devices, wherein, In the equalization control process, the voltage, current and charge state changes of the plurality of battery cell module devices are continuously monitored, and the equalization strategy parameters are dynamically adjusted so as to simulate the equalization process among the battery cells in the real battery pack.
8. The method of claim 1 or 5, wherein during the real-time simulation, the method further comprises: Setting overvoltage, overcurrent, short circuit and overtemperature thresholds in the cell module equipment, and monitoring the output end state in real time through an acquisition circuit, and When the output voltage, current or temperature parameter is detected to exceed a preset threshold value, triggering a protection control instruction to stop outputting or enter a controlled load-reducing state, and simultaneously transmitting abnormal state information to a target terminal.
9. Cell module testing arrangement of fusion regulatory function, characterized in that includes: The simulation terminal downloads a parameter setting instruction and transmits real-time parameter data of each single battery cell of the battery cell simulation equipment; The cell module comprises a microprocessor, a digital-to-analog conversion circuit, an acquisition circuit, an operational amplifier buffer, a signal comparator, a signal integrator, a power operational amplifier circuit, a sampling resistor, a communication interface, a Sense interface and a VBAT electric quantity output interface, and A control unit for controlling and executing the cell core cell testing method of any of claims 1-9, comprising: the data processing module is used for generating multidimensional cell characteristic parameters according to experimental data and historical operation data of the target cell in different working phases; the model construction module is used for constructing a cell data model capable of reflecting the dynamic behavior of the cell by adopting a machine learning method or a statistical modeling method according to the multidimensional cell characteristic parameters; The simulation control module is used for controlling the cell module to output the simulated cell voltage, current and charge state change according to the cell equivalent circuit model so as to realize real-time simulation of the target cell under different charge and discharge working conditions, and the simulation control module is used for controlling the cell module to output the simulated cell voltage, current and charge state change according to the cell equivalent circuit model And the dynamic correction module is used for acquiring preset data of the output end of the cell simulation equipment in real time in the real-time simulation process and carrying out dynamic correction and parameter adjustment on the cell data model according to the preset data acquired in real time.
10. An electronic device, comprising: a processor adapted to execute a computer program, and A computer readable storage medium having stored therein a computer program which, when executed by the processor, implements the method of any one of claims 1 to 8.

Description

Cell module testing method and device and electronic equipment Technical Field The present application relates to the field of cell testing technology, and more particularly, to a cell testing method, apparatus, electronic device, and computer-readable storage medium. Background With the rapid development of new energy automobiles, energy storage systems and related new energy industries, a battery system is used as a key energy unit, the performance, the safety and the service life of the battery system become core factors for limiting the stability of the battery system, and a battery core is a core unit in the battery system, and the characteristics of voltage, current, charge state change, temperature and the like of the battery system directly influence the control strategy and the operation safety of the battery system, so that the performance characteristics of the battery core under various charge and discharge working conditions, environmental conditions and load conditions are usually required to be tested, analyzed and verified in the development, design and verification stages of the battery system so as to evaluate the performance and provide a basis for the formulation of a control strategy. At present, the test of the performance of the battery cell still mainly depends on the charge and discharge experiment mode of the actual battery cell, namely, the actual battery cell is used for carrying out multi-round charge and discharge test in an experiment bench or a battery pack, so that the operation data of the battery cell under different working conditions are obtained, and although the mode can obtain relatively real test results, in the actual application process, on one hand, the charge and discharge experiment of the actual battery cell generally needs a relatively long test period and is difficult to meet the requirements of rapid iteration and rapid research and development of products, on the other hand, the number of the battery cells needed by the experiment is large, the cost is high, the test process has relatively high requirements on the experiment environment and safety protection, and particularly, under the test scene of a high-power and high-capacity battery pack, the safety risks such as overheat and overcurrent are easy to occur, so that the experiment cost is increased, and the repeated verification on complex working conditions and extreme working conditions under the laboratory environment is limited. Disclosure of Invention The application provides a cell module testing method, a device and electronic equipment, which can integrate data driving modeling and equivalent circuit modeling and introduce a dynamic adjustment and model correction mechanism in the simulation process so that a cell module result can continuously approach to the real state of a target cell along with the change of working conditions. According to the method, according to experimental data and historical operation data of a target cell in different working stages, multidimensional cell characteristic parameters are generated, a cell data model capable of reflecting dynamic behaviors of the cell is built by adopting a machine learning method or a statistical modeling method according to the multidimensional cell characteristic parameters, the cell data model is mapped into an equivalent circuit model, a cell equivalent circuit model is generated by setting preset circuit parameters, and cell analog module equipment is controlled to output analog cell voltage, current and state of charge changes according to the cell equivalent circuit model, so that real-time simulation of the target cell under different charge and discharge working conditions is achieved. In an alternative of the first aspect, the method includes obtaining experimental data and historical operating data of a target cell in different working phases, wherein the experimental data includes voltage, current, temperature and state change data of the target cell in different working phases, the historical operating data includes voltage, current, temperature and state change data of a historical record of the target cell, after the experimental data and the historical operating data are obtained, data denoising, abnormal data removing, time alignment and normalization processing are performed, and multidimensional feature parameters for representing voltage change rate, current response feature, temperature change feature, state and internal resistance change of the cell state are extracted, the method includes obtaining experimental data and historical operating data of the target cell in different working phases, wherein the experimental data includes voltage, current, temperature and state change data of the target cell in different state of charge, the historical operating data include voltage, current, temperature and state change data of the target cell historical record are set as an equivalent circuit, the equivalent circuit is s